Working Papers
The Impact of Economic Shocks on Financial Access: Evidence from COVID-19 Pandemic
Conferences: FMA 2021, SFA 2021, New Zealand Financial Market Conference 2021, AFA Poster Session 2022, SWFA 2022
- This paper studies individual borrowing activity through small-dollar loans from non-depositary consumer lenders.
We exploit the COVID-19 economic shock and the subsequent government response to determine the factors affecting the demand for small-dollar loans using foot traffic to lender locations.
We find that the number of visitors to lenders drops significantly following shelter-in-place orders and government relief programs.
The results show that the lockdowns suppressed financially underserved consumers’ access to credit. The supplemental paychecks program helped reduce consumers’ need for credit.
We find differential impacts of government relief programs in metropolitan areas and bank desserts, and on bank borrowers and borrowers from non-depository lenders.
Reassessing Firms' Environmental Ethics and Impact: An Efficiency-Based Carbon Pricing Approach
Coauthored with Stefano Bonini and Meghana Vaidya, the latest version
Conferences: Semi-finalist for the Best Paper Award in FMA 2022, 8th International Symposium on Environment & Energy Finance Issues
- We model the firm's objective as a function of output and environmental ethics. The cost of emission increases with production and is weighted by firms’ environmental ethics, leading firms to endogenize the optimal emission-output level. Firms with higher environmental ethics have higher marginal output and emit less because of the higher emission cost. More importantly,
we argue that the one-size-fits-all carbon pricing is not optimal. Instead, carbon emissions should be priced based on the
efficiency of the emission. Given a fixed carbon cap, switching to efficiency-based carbon pricing increases social
welfare. Using emission data from 1995 to 2020, we provide empirical evidence to support the theory.
Technology Spillover, Bank Branch Closings, and Small Business Lending
Solo paper, draft available upon request, slides here
- We study the impact of bank technology on credit supply when banks close branches. Although banks reduce small business lending in counties where they close branches, the reduction is less for banks with superior technology.
The effect is more pronounced among high-income borrowers and in areas where closed branches have served longer. We do not find a significant impact of technology on small business lending when banks expand branches.
Our results suggest that bank branches remain important in providing credit to the local communities.
Technology helps preserve the existing customer-bank relationship but cannot reduce information asymmetry.
Work in Progress
Investor Overconfidence and Learning: Evidence from P2P
Coauthored with Anand Goel and Hongju Ren
We use a large microloan dataset from a leading Peer-to-Peer (P2P) lending platform in China to study the behavior of P2P investors. We find that investors who actively choose loans face a higher probability of loan default and earn a lower return than investors who delegate investment decisions to the platform.
We show that overconfidence is a plausible explanation for the behavior of these investors. We provide evidence of investors' learning behavior: investors who experience a loan default are likely to select less risky loans and earn a higher return in the future.
Investors' Attention Constraints and Information Presentation
Coauthored with Anand Goel and Hongju Ren
We draw on microloan data from a P2P platform in China to study how P2P's platform design can impact investors' decisions. The results show that the investment performance of investors worsens when they use a smartphone app rather than a computer, suggesting that information content and presentation impact the quality of investment decisions.
Besides, mobile app users are less likely to choose a low-interest-rate project than computer users after experiencing loan default.
The results imply that the cross-sectional difference in return and learning behavior of mobile app and computer users may stem from attention constraints.
